Methylnaltrexone (MNTX) is a quaternary derivative of the pure opioid antagonist, naltrexone. It exists as a salt. Names used for the bromide salt of MNTX in the literature include: Methylnaltrexone bromide; N-Methylnaltrexone bromide; Naltrexone methobromide; Naltrexone methyl bromide; MRZ 2663BR. MNTX was first reported in the mid-70s by Goldberg et al as described in U.S. Pat. No. 4,176,186. It is believed that addition of the methyl group to the ring nitrogen forms a charged compound with greater polarity and less liposolubility than naltrexone. This feature of MNTX prevents it from crossing the blood-brain barrier in humans. As a consequence, MNTX exerts its effects in the periphery rather than in the central nervous system with the advantage that it does not counteract the analgesic effects of opioids on the central nervous system.
MNTX is a chiral molecule and the quaternary nitrogen can be in R or S configuration. (See FIG. 1.) It is unknown whether the different stereoisomers of MNTX exhibit different biological and chemical properties. All of the reported functions of MNTX described in the literature indicate that MNTX is a peripheral opioid antagonist. Some of these antagonist functions are described in U.S. Pat. Nos. 4,176,186, 4,719,215, 4,861,781, 5,102,887, 5,972,954, 6,274,591, 6,559,158, and 6,608,075, and in U.S. patent application Ser. No. 10/163,482 (2003/0022909A1), Ser. No. 10/821,811 (20040266806), Ser. No. 10/821,813 (20040259899) and Ser. No. 10/821,809 (20050004155). These uses include reducing the side-effects of opioids without reducing the analgesic effect of opioids. Such side-effects include nausea, emesis, dysphoria, pruritus, urinary retention, bowel hypomotility, constipation, gastric hypomotility, delayed gastric emptying and immune suppression. The art discloses that MNTX not only reduces the side-effects stemming from opioid analgesic treatment but also reduces the side-effects mediated by endogenous opioids alone or in conjunction with exogenous opioid treatment. Such side-effects include inhibition of gastrointestinal motility, post-operative gastrointestinal dysfunction, idiopathic constipation and other such conditions including, but not limited to, those mentioned above. However, it is unclear from the art whether the MNTX used in these studies was a mixture of R and S stereoisomers or a single stereoisomer.
The art suggests that isolated stereoisomers of a compound sometimes may have contrasting physical and functional properties, although it is unpredictable whether this is the case in any particular circumstance. Dextromethorphan is a cough suppressant, whereas its enantiomer, levomethorphan, is a potent narcotic. R,R-methylphenidate is a drug to treat attention deficit hyperactivity disorder (ADHD), whereas its enantiomer, S,S-methylphenidate is an antidepressant. S-fluoxetine is active against migraine, whereas its enantiomer, R-fluoxetine is used to treat depression. The S enantiomer of citalopram is therapeutically active isomer for treatment of depression. The R enantiomer is inactive. The S enantiomer of omeprazole is more potent for the treatment of heartburn than the R enantiomer.
Bianchetti et al, 1983 Life Science 33 (Sup I):415-418 studied three pairs of diastereoisomers of quaternary narcotic antagonist and their parent tertiary amines, levallorphan, nalorphine, and naloxone, to see how the configuration about the chiral nitrogen affected in vitro and in vivo activity. It was found that the activity varied considerably depending on how the quaternary derivatives were prepared. In each series, only the diastereomer obtained by methylation of the N-allyl-substituted tertiary amine (referred to as “N-methyl diastereomer”) was potent in displacing 3H-naltrexone from rat brain membranes, and acting as a morphine antagonist in the guinea-pig ileum. Conversely, diastereoisomers obtained by reacting N-methyl-substituted tertiary amines with allyl halide (referred to as “N-allyl diastereomers”) did not displace 3H-naltrexone and had negligible antagonist activity and slight agonist action in the guinea-pig ileum. In vivo findings were generally consistent with those in vitro. Thus only the “N-methyl” but not the “N-allyl diastereomers” inhibited morphine-induced constipation in rats and behaved as antagonists. The author stated that the prepared materials appeared to be pure by 1H and 13C nuclear magnetic resonance (NMR) analysis, but these methods are not accurate. The author cites a literature reference for the assignment of the R configuration to the “N-methyl diastereomer” of nalorphine. No assignment is proposed for the levallorphan and naloxone diastereomers. It would be adventurous to extrapolate the configuration to these diastereomers (R. J. Kobylecki et al, J. Med. Chem. 25, 1278-1280, 1982).
Goldberg et al.'s U.S. Pat. No. 4,176,186, and more recently Cantrell et al.'s WO 2004/043964 A2 describe a protocol for the synthesis of MNTX. Both describe a synthesis of MNTX by quatemizing a tertiary N-substituted morphinan alkaloid with a methylating agent. Both Goldberg et al. and Cantrell et al. are silent as to the stereoisomer(s) produced by the synthesis. The authors remained cautiously silent about the stereochemistry because the stereochemistry could not be determined based on prior art. The cyclopropylmethyl side-chain in naltrexone is different from the prior art side-chains and may have affected the stereochemical outcome in the synthesis of MNTX, as may other reaction parameters such as temperature and pressure. Based on the method of synthesis described in each, it is unknown whether the MNTX so produced was R, S or a mixture of both.
S-MNTX in pure form, and a method of making pure S-MNTX have not been described in the literature. Researchers would have been unable to definitively characterize and distinguish the stereoisomer(s) obtained by the Goldberg et al. or Cantrell et al. synthesis in the absence of S-MNTX as a standard.